We propose to evaluate the effectiveness of a newly established needle exchange program (NEP) to be operated by the Baltimore City Health Department (BCHD) in a stationary site with up to 1,000 slots. The population for evaluation involves all NEP participants and subjects in three ongoing community based cohort studies of HlV infection among injecting drug users (lDUs) in Baltimore (i.e., ALIVE, HERS, New Initiates). The ALIVE Study is following 2,000 IDUs and is reopening recruitment for another 400 IDUs. The HERS and New Initiates study has open recruitment for another 600 IDUs. For each of these cohorts we will ascertain who joins and who avoids the NEP. NEP participants not already enrolled will be encouraged to join one of these studies. NEP participants who choose not to be subjects in these other studies will be compared to both NEP and non-NEP participants who newly enroll in these other studies. All participants in the NEP and each of the cohort studies will undergo interviews on demographics, drug use and sex practices, knowledge and attitudes about HIV and prevention practices, and utilization of social services including treatment for drug abuse. All subjects will have serologies performed semiannually. In addition, returned needles of NEP participants are linked to individuals and a sample will undergo biological assays for HIV and protein/DNA phenotypes to validate previous models which were based only upon analysis of returned needles. For analyses, we will first compare characteristics of IDUs who join versus those who avoid the NEP, stratified by cohort membership, using regression procedures. We will also compare NEP participants who are members or are not members of the cohort studies. Second, we will compare temporal patterns of drug use and sex practices by membership in the NEP stratified by cohort using longitudinal data analytic techniques. Third, we will examine rates of HIV seroconversion by NEP status using person-time techniques. Nested case-control analyses of HIV seroconverters and seronegative individuals will be performed to ascertain more detailed information on risk and practice behaviors including dynamics of needle use. These analyses will provide important information to determine whether NEP is associated with lower levels of risk behaviors and reduction in the rates of new HIV infection.